The frontline support team will be unavailable to answer questions on April 15th and 17th 2019. We will be back soon after. Thank you for your patience and we apologize for any inconvenience!

HaplotypeCaller crash with multithreading + SSE

I recently ran the HaplotypeCaller using both the new --pairHMM options and -nct 2, and I encountered a crash in the program (8 of 96 runs crashed). I have rerun the analysis without the "-nct 2" option, and it appears to work as expected, so I'm not in a hurry. I am currently using GATK 3.1-1; below is the specific command I used as well as the stack trace that resulted.

GenomeAnalysisTK -T HaplotypeCaller -R human_g1k_v37_decoy.fasta -I [email protected] --dbsnp dbsnp_137.b37.vcf.gz -L targets.GRCh37.bed -A QualByDepth -stand_emit_conf 10 -mbq 20 --downsample_to_coverage 300 -A HaplotypeScore -A MappingQualityRankSumTest -A ReadPosRankSumTest -A FisherStrand -A GCContent -A AlleleBalanceBySample -A AlleleBalance -o HG00593.vcf.gz -sample_rename_mapping_file rename_file -nct 2 -ERC GVCF -pairHMM VECTOR_LOGLESS_CACHING -variant_index_type LINEAR -variant_index_parameter 128000

Stack trace:

ERROR stack trace

at java.util.LinkedHashMap$LinkedHashIterator.nextEntry(
at java.util.LinkedHashMap$
at java.util.LinkedHashMap$
at org.broadinstitute.sting.gatk.walkers.haplotypecaller.GenotypingEngine.addMiscellaneousAllele(
at org.broadinstitute.sting.gatk.walkers.haplotypecaller.GenotypingEngine.assignGenotypeLikelihoods(
at org.broadinstitute.sting.gatk.traversals.TraverseActiveRegions$TraverseActiveRegionMap.apply(
at org.broadinstitute.sting.gatk.traversals.TraverseActiveRegions$TraverseActiveRegionMap.apply(
at org.broadinstitute.sting.utils.nanoScheduler.NanoScheduler$
at java.util.concurrent.Executors$
at java.util.concurrent.FutureTask$Sync.innerRun(
at java.util.concurrent.ThreadPoolExecutor.runWorker(
at java.util.concurrent.ThreadPoolExecutor$

If you require anything else from me, please let me know.

Best Answer


Sign In or Register to comment.